Chapter 1
Translational Cancer Research: Gene Therapy by Viral and Non-viral Vectors

Vincenzo Cerullo, Kilian Guse, Markus Vähä-Koskela, and Akseli Hemminki

University of Helsinki, Helsinki, Finland

Adenovirus

Adenovirus is among the most used vectors for gene therapy and gene transfer, and about 23% of all vector-based clinical trials have been performed with it (www.wiley.com//legacy/wileychi/genmed/clinical/). Adenovirus was first isolated in 1953 from human adenoids [1]. To date, 55 different human serotypes, subdivided into seven subgroups (A–G), have been characterized [2,3].

Adenovirus is a nonenveloped double-stranded DNA virus surrounded by an icosahedral protein capsid (Table 1.1). The capsid comprises penton and hexon proteins with knobbed fibers protruding from the vertices of the capsid [4]. Soon after its entry into the target cell viral DNA reaches the nucleus where starts its replication. Early genes, mainly involved in DNA replication, are transcribed first [5], followed by late genes mainly coding for structural proteins [4].

Table 1.1 The main characteristics of the viruses discussed in this chapter.

• Replication-competent vectors

• First-generation adenoviral vectors

• Helper-dependent adenoviral vectors

• Deletion of genes essential for replication in normal cells (E1A, E1B)
• Tumor-specific promoters
• MicroRNA targets in genome

• Deletion of genes essential for replication in normal cells (VGF, TK)
• MicroRNA targets in genome

• Deletion of genes essential for replication in normal cells (TK, RR, γ34.5)
• Tumor-specific promoters
• MicroRNA targets in genome

• Tumor-specific promoters

• scAb-binding domain in knob
• Cell-specific peptides in fiber/knob
• Serotype fiber/knob exchange

• scAb-binding domain in knob
• Cell-specific peptides in fiber/knob
• serotype fiber/knob exchange

• scAb-binding domain in knob
• Cell-specific peptides in fiber/knob
• Serotype fiber/knob exchange

dsDNA, double-stranded DNA; HSV, herpes simplex viruse; MVA, modified vaccinia Ankara; ssDNA, single-stranded DNA; TK, thymidine kinase; VGF, vaccinia growth factor.

Adenoviruses tend to be species-specific with regard to permissivity to replication. However, there may be some exceptions to this general rule. It has been reported that adenovirus serotype 5 subgroup C (usually referred as Ad5, the most used gene therapy vector) can replicate to some degree in cotton rats [6,7], New Zealand rabbits [8], and Syrian hamsters [9]. This feature of Ad5 has been very important for scientists around the world because it has allowed them to use these animal models to develop new therapies for disease.

Historically adenovirus has been the most used vector for gene therapy and gene-transfer purposes. In 1970s F. Graham and colleagues discovered the importance of the E1 gene, that made possible the use of adenovirus as a viral vector for gene therapy [10]. In fact, as E1 gene products initiate the replication of the viral DNA, serotype 5 adenoviruses with E1 deleted are incapable of replicating and remain episomal. Taking advantage of this characteristic, scientists replaced E1 with different expression cassettes to avoid virus replication while promoting expression of the transgene inserted in place of E1. Later on, E1-deleted adenoviral vectors, also known as first-generation adenoviral vectors (FG-Ad), were developed into high-capacity adenoviral vectors or Helper-dependent adenoviral vectors (Hd-Ad). HD-Ad are devoid of all viral genes except the two inverted terminal repeats (ITRs) and the packaging signal (psi). They show a high cloning capacity (up to 36 kb) and reduced immunogenicity and toxicity [11] (Figure 1.1). Since then, it has been mainly used as vector for gene transfer for genetic diseases [12] or to treat cancer [13]. The immunogenicity of adenovirus may render it unsuitable for long-term gene expression but makes it attractive for treatment of cancer. Use of a replication-deficient adenovirus as a gene delivery vehicle is the classic approach, with some exciting clinical results [14,15,16,17], but no products have been approved outside of China. This approach has been reviewed recently [18]. In the past decade, many adenoviral gene therapists have focused on use of adenovirus as a replication-competent oncolytic virus and thus this will be focus of this chapter.

Figure 1.1 Schematic diagram representing the different kinds of adenovirus-derived vectors used for gene therapy. (A) Wild-type adenovirus is able to replicate and kill all permissive cells. (B) The E1 gene is replaced by the expression cassette; this vector can infect all permissive cells but they cannot replicate unless E1 is not transcomplemented by the packaging cell line. (C) All viral genes are deleted except ITRs and the packaging signal. These vectors can infect all permissive cells but they cannot replicate. (D) Oncolytic adenoviruses. These viruses have been engineered to selectively replicate in and kill cancer cells.

Oncolytic Adenoviruses for Treatment of Cancer

Oncolytic adenoviruses are specifically modified to selectively replicate in and destroy cancer cells. This selectivity is achieved by modifications of the genes involved in viral replication so that the life cycle of the virus can occur only in cells than can transcomplement the defect, including cancer cells, while the replication of the virus is arrested in normal cells (transcriptional targeting) (Figure 1.2). An alternative approach is to use tumor-specific promoters to “drive” E1 expression to allow selective replication of the virus in cancer cells [19] (Figure 1.2).

Figure 1.2 Transcriptional targeting. Simplified schematic illustrating the strategies used to achieve transcriptional targeting of tumor cells. (A) For example, a viral genome is modified to not be able to counteract the defense mechanisms that a normal cell turns on following a viral infection. (B) Tumor cells are defective of such mechanisms hence the virus can have its normal life cycle. (C) Tumor-specific promoter can initiate virus DNA replication, starting its life cycle.

Historically, the first adenoviruses used in patients were wild-type viruses [20]. The concept was revived with the first adenovirus proposed to have tumor selectivity, dl1520 (today known as ONYX-015) [21]. This adenovirus bears a naturally occurring variation that results in a nonfunctional E1B-55k product. E1B-55k is one of the proteins encoded by the early gene E1 and its normal function is to promote the degradation of p53 to avoid the infected cell undergoing apoptosis [22]. In infected normal cells p53 is not degraded by the mutated E1B-55k so that they can smoothly continue towards cell cycle arrest and apoptosis, which causes the arrest of the virus’s life cycle; on the other hand, in cancer cells, where the p53/p14ARF pathway is universally defective, the mutation is not needed to avoid apoptosis [21]. An issue with this type of virus is that E1B-55k is needed for late mRNA transport and its absence results in ineffective oncolysis, several orders of magnitude less than with the wild-type virus [23].

An alternative strategy used to generate adenoviruses selective for cancer cells is a 24 bp deletion of the E1A gene [23,24,25]. This deletion results in the inability of E1A to bind to retinoblastoma tumor-suppressor protein (Rb) and to release eukaryotic initiation factor E2F, which in the case of wild-type adenovirus would result in S-phase induction in normal cells. Therefore the “delta-24” viruses are unable to induce S-phase in host cells and no viral replication follows. In contrast to normal cells, most if not all cancer cells have a defective Rb/p16 pathway, rendering the...